Little is known about which psychological and social factors affect decisions about when to purchase and consume alcohol. The purpose of the research described in this proposal is to determine some of the subjective and social factors which influence these choices. The alcohol availability literature suggests that physical and economic conditions which make alcohol more difficult to obtain (e.g., restricted hours of sale) reduce alcohol consumption (Beauchamp, 1981; Smart, 1977b). Smart (1977a) argues that the availability construct should be expanded to include subjective and social components. Subjective availability concerns individuals' perceptions of how accessible alcohol is to them. Social availability concerns how accessible alcohol is within one's social network. It is hypothesized that people who perceive alcohol to be easily available will buy more alcohol and drink more than those who do not. Individuals who feel that alcohol outlets are conveniently located, who do not mind going out of their way to buy alcohol, who feel no discomfort about entering liquor stores or bars, who stock up on alcohol so it is easily available at home, and who feel that alcohol is reasonably priced are hypothesized to experience high levels of subjective availability and consequently to buy and consume more alcohol than individuals who do not perceive alcohol to be easily available. Individuals who have friends, family, and coworkers who approve of drinking alcohol, who make alcohol available at parties, and who encourage drinking by virtue of their own behavior are hypothesized to experience high levels of social availability and consequently to buy and consume more alcohol than individuals who do not perceive alcohol to be easily available. It is also hypothesized that subjective availability will moderate the effects of physical availability (distance from an outlet) on consumption. These hypotheses will be examined in a survey of 780 Michigan residents from eight counties. Counties were selected to vary in alcohol outlet density (high, medium, low). Thirty minute telephone interviews will be conducted. Multiple indicators of most concepts will be included. Multiple regression and structural modeling techniques will be used to test the hypothesized relationships between physical, subjective, and social availability, and consumption. Alcohol abuse is a serious national health problem. This study's results could lead to the design of alcohol prevention programs specifically targeted at high risk individuals.